Running Head: PTSD COGNITIONS & ATTACHMENT IN VICTIMS. PTSD-related Cognitions and Romantic Attachment Style as Moderators of

Transcription

1 Running Head: PTSD COGNITIONS & ATTACHMENT IN VICTIMS PTSD-related Cognitions and Romantic Attachment Style as Moderators of Psychological Symptoms in Victims of Interpersonal Trauma Lisa S. Elwood & Nathan L. Williams University of Arkansas In Press: Journal of Social and Clinical Psychology Correspondence to: Lisa S. Elwood, Department of Psychology 216 MEMH, University of Arkansas, Fayetteville, AR Telephone: (479) ; Fax: (479) ;

2 PTSD Cognitions & Attachment in Victims 2 Abstract Both PTSD-related cognitions and adult romantic attachment cognitions have been linked to level of symptomatology and can be conceptualized as interpersonal cognitive schemas that may increase vulnerability to the development of psychological symptoms. The present study examined PTSD-related cognitions and adult romantic attachment as moderators of symptom development after interpersonal trauma. Victims reported higher levels of PTSD-related cognitions and attachment anxiety, but not attachment avoidance, than non-victims. Results suggest that the experience of trauma, high levels of insecure attachment, and high levels of PTSD-related cognitions are associated with higher levels of symptomatology. In addition, PTSD-related cognitions served as a unique predictor of PTSD symptoms, and moderated the relation between attachment anxiety and depressive symptoms. Keywords: PTSD, cognitions, attachment, interpersonal trauma 2

3 PTSD Cognitions & Attachment in Victims 3 PTSD-related Cognitions and Romantic Attachment Style as Moderators of Psychological Symptoms in Victims of Interpersonal Trauma Rates of the prevalence of rape and physical abuse in college populations are generally similar to or higher than those found in the general population (Green et al., 2000; Kirk & Dollar, 2002; Scarpa et al., 2002; Vrana & Lauterbach, 1994; Kessler, Sonnega, Bromet, Hughes, & Nelson, 1995). While PTSD is the disorder most commonly associated with the experience of traumatic events such as interpersonal trauma, other disorders afflict victims as well. The common comorbidity of PTSD suggests that disorders such as depression or other anxiety disorders may be part of the aftermath of trauma (Resick, 1993; Orsillo et al., 1996). Given both the high prevalence of interpersonal trauma in the college student population and the likelihood that exposure to such trauma may lead to the development of emotional disorders and/or symptoms, it seems critical to assess vulnerability factors and moderators of psychological symptoms in this population. Cognitive-behavioral theories of PTSD describe individual differences that may function as vulnerability factors to the development of PTSD. Emotional processing theory (EPT) suggests that many negative events in life are emotionally re-experienced, but that normally the frequency and intensity decreases over time (i.e., habituation; Foa & Rothbaum, 1998). A person s pre-trauma schemas, memory of the event, and memory of experiences prior to the event are posited to interact and impede the emotional processing of the trauma (Foa & Riggs, 1993). The failure of the habituation process is proposed to lead to the development and maintenance of PTSD symptoms (Foa & Rothbaum, 1998). Ehlers and Clark s (2000) cognitive model of PTSD proposes that PTSD develops and 3

4 PTSD Cognitions & Attachment in Victims 4 persists when individuals process a past trauma in a way that leads to the sense of a serious, current threat. The erroneous perception of an event as constant and ongoing may occur as the result of a maladaptive cognitive processing of the event and a disturbance in the memory of the event (Ehlers & Clark, 2000). According to these theories, being able to accept the trauma as an unfortunate random event, instead of a reflection of the norm, is essential to protect oneself from developing or maintaining PTSD symptoms. In a further attempt to identify cognitive vulnerabilities to PTSD, Foa and colleagues (Foa, Ehlers, Clark, Tolin, & Orsillo, 1999) attempted to empirically identify the cognitive beliefs most related to the development of PTSD. Consideration of theoretical conceptualizations and clinical experience resulted in the development of the Posttraumatic Cognitions Inventory, which assesses three domains of relevant cognitions: a) negative cognitions about the self, b) negative cognitions about the world, and c) selfblame for the trauma. Although admittedly a broad term, the current study will retain the term used in the original article, PTSD-related cognitions, when referring to these dimensions. These three cognitive belief domains have been found to be predictive of PTSD severity, depression, and anxiety in traumatized individuals and to discriminate between victims with and without PTSD (Foa et al., 1999). Adult Romantic Attachment Dimensions One vulnerability factor that has frequently been examined in relation to both interpersonal trauma and PTSD is attachment. Attachment originally focused on the relationships between children and their caregivers, with the assumption that these relationships would influence an individual s later expectations and behaviors in other relationships (Bowlby,1973). Based on the parent-child attachment theory, the 4

5 PTSD Cognitions & Attachment in Victims 5 development of an insecure attachment is proposed as a common consequence of child abuse (Alexander, 1992). Attachment styles have been found to be relatively stable across a person s lifespan and thus to guide the individual s contact with others into adulthood (Main, Kaplan, & Cassidy, 1985). Recently the study of attachment has expanded to include adult romantic relationships. Romantic attachment styles can be thought of as interpersonal schemata, or internal working models of romantic relationships, that can be used to describe an individual s beliefs about the availability and responsiveness of romantic partners and to predict their behavior in interpersonal relationships (Brennan, Clark, & Shaver, 1998). The two-dimensional approach to adult romantic attachment examines the presence of anxiety related to fears of abandonment and avoidance of romantic relationships, based on individuals models of self and others (Brennan et al., 1998). The attachment avoidance dimension examines a person s discomfort with closeness and dependency, while the anxiety dimension encompasses a person s anxiety related to abandonment. Adults (McCarthy & Taylor, 1999; Muller & Lemieux, 2000; Muller, Sicoli, & Lemieux, 2000, Roche, Runtz, & Hunter, 1999; Twaite & Rodriguez-Srednicki, 2004; Whiffen, Judd, & Aube, 1999) and adolescents (Shapiro & Levendosly, 1999) endorsing the experience of childhood abuse have consistently reported high levels of insecure attachment. In addition, high levels of insecure attachment have been associated with increased distress and symptomatology in adults (Muller & Lemieux, 2000), and older children and adolescents (Sternberg, Lamb, Guterman, Abbott, & Dawud-Noursi, 2005). The relation between childhood abuse and insecure attachment has been found in studies using both categorical (Muller & Lemieux, 2000; Roche, Runtz, & Hunter, 1999; Shapiro 5

6 PTSD Cognitions & Attachment in Victims 6 & Levendosky, 1999) and dimensional (McCarthy & Taylor, 1999; Muller & Lemieux, 2000; Twaite & Rodriguez-Srednicki, 2004; Whiffen, Judd, & Aube, 1999) models of attachment. Several studies have tested a mediational model in which attachment mediates the relation between child abuse and the development of psychological symptoms. Several studies have found support for adult attachment as a mediator between reported childhood abuse and adult psychological symptoms (Roche et al., 1999; Shapiro & Levendosky, 1999; Twaite & Rodriguez-Srednicki, 2004). However, one study found support for a moderational, rather than mediational, model, indicating that attachment moderated between the experience of child sexual abuse and the development of depressive symptoms (Whiffen et al., 1999). In addition to PTSD, insecure attachments have been linked to higher levels of a variety of anxious (Cassidy, 1995; Williams & Riskind, 2004) and depressive symptoms (Haaga et al., 2002; Williams & Riskind, 2004). A stable negative attachment style can make individuals vulnerable to symptom development across the life span. Individuals with insecure attachments are believed to be especially vulnerable to symptom development after negative experiences. For example, attachment has served as a significant predictor of the onset and severity of depression and anxiety after stressful life events (Hammen, Burge, Daley, Davila, Paley, and Rudolph, 1995). The current study examines previously established, presumably stable, attachment styles as vulnerability factors for the development of emotional symptoms. Previous studies have examined attachment as a mediator between child abuse and the development of PTSD. Mediational models are useful when examining childhood attachment and abuse, due to the abuse occurring during a developmental period for the 6

7 PTSD Cognitions & Attachment in Victims 7 attachment. Victims of adult interpersonal trauma are presumed to have established, stable attachment styles. Therefore, the experience of adult interpersonal trauma is not believed to have a causal influence on the individual s attachment. The lack of a direct causal pathway between the experience of interpersonal trauma and attachment suggests that attachment be examined as a moderator, rather than mediator, between the trauma and the symptom development (Baron & Kenny, 1986). Present Study The present study expands upon previous research by examining adult romantic attachment cognitions and the effects of adult interpersonal trauma. It is proposed that one s adult romantic attachment will influence his/her interpretation of and reactions to interpersonal trauma. A person with an insecure adult romantic attachment may interpret and react to the event negatively, such as believing that they are responsible for the event and that people are not trustworthy. Alternatively, a secure adult romantic attachment may allow the individual to interpret and respond more positively, such as recognizing the unpredictability and uniqueness of the event. A moderational model is proposed in which adult romantic attachment partially moderates between the experience of interpersonal trauma and the development of psychological symptoms in young adult college students. The present study also examines the effects of PTSD-related cognitions on the development of psychological symptoms after experiencing interpersonal trauma. The study aims to replicate and extend the results found by Foa and colleagues (1999) which suggest that PTSD cognitions are critical to the development of PTSD and are concurrently related to increased levels of psychological symptoms. The present study 7

8 PTSD Cognitions & Attachment in Victims 8 examines the influence of these two proposed vulnerability factors by examining the independent relations between each factor and level of symptomatology. Finally, the present study examines the influence of the interaction between attachment cognitions and PTSD-related cognitions on the development of the psychological symptoms. As both adult romantic attachment and PTSD-related cognitions are cognitive schemas about the self in relation to others, it is possible that the two may interact to influence the development of PTSD. As previously described, adult romantic attachment is reflective of the individual s comfort with and expectations about interpersonal relationships based on beliefs about the self and others. PTSD related cognitions, on the other hand, describe individuals beliefs about themselves and the general world. Therefore, while adult attachment styles focuses on beliefs about known others in relation to the self and expectations for ongoing relationships, PTSD related cognitions describe beliefs about the self, unknown others, and expectations for interactions with people with whom the individual may not have ongoing relationships. PTSD related cognitions and attachment style could interact in several ways. First, individuals that hold negative beliefs about known others may be at an increased likelihood to develop negative beliefs about unknown others. Second, an individual s attachment style is likely to influence his or her beliefs about the cause of the experience and integration of the experience into their beliefs about the world. An individual with an insecure attachment is predisposed to distrust others and/or criticize his or her self. This tendency to view others and the self negatively is likely to be exacerbated after the interpersonal trauma and may result in a generalized distrust of others and self-blame (i.e., PTSD related cognitions). 8

9 PTSD Cognitions & Attachment in Victims 9 Three hypotheses were tested in this study. 1) Attachment dimensions were proposed to moderate the relation between interpersonal trauma and the resulting psychological symptoms (i.e., PTSD, anxious, and depressive symptoms). 2) PTSD-related cognitions (e.g., negative cognitions about the self, negative cognitions about the world, and self-blame) were expected to moderate the relation between interpersonal trauma and resulting psychological symptoms. 3) Attachment dimensions were expected to moderate the relation between PTSD-related cognitions and level of psychological symptoms. Method Participants & Procedure Participants were 287 undergraduate students at a large Southwestern Public University. Participants were 76% female and 89% Caucasian (4% African American, 2% Asian, 1% Hispanic, and 3% other). The majority of the participants were freshmen or sophomores, with a mean age of (SD = 3.08). Participants were recruited from a psychology undergraduate research pool, assembled in small groups, and asked to complete the following questionnaires. Measures Beck Anxiety Symptom Inventory (BAI). The BAI (Beck, Epstein, Brown, & Steer, 1988) is a 21-item measure designed to assess the severity of anxious symptoms. Participants rate how much various symptoms of anxiety have bothered them during the past week on a 4-point Likert scale, with answers ranging from not at all to I could barely stand it. The internal consistency of the BAI in this study was excellent (α =.92). 9

10 PTSD Cognitions & Attachment in Victims 10 Beck Depression Inventory-II (BDI-II). The BDI-II (Beck, Steer, Brown, 1996) is a 21-item measure that assesses levels of depressive symptoms. Respondents read a group of four statements and choose the one that best describes how they have felt during the past week (0 to 3). The BDI-II is a reliable and well-validated measure of depressive symptoms (Beck, Steer, & Garbin, 1988). The internal consistency of the BDI-II in this study was excellent (α =.92). Experiences in Close Relationships Scale (ECR). The ECR (Brennan et al., 1998) scale is a 36-item questionnaire that measures adult romantic attachment style. Respondents are given a description of a possible attitude towards relationships, such as I worry about being abandoned, and asked to rate them on a 5-point Likert scale, ranging from disagree strongly to agree strongly. The ECR contains 2 subscales that measure orthogonal dimensions of adult romantic attachment: attachment avoidance and attachment anxiety. The authors examined the two dimensions in relation to 14 popular attachment measures, including Griffin and Bartholomew s (1994), Simpson s (1990), and Collins and Read s (1990) scales, and generally found the expected relations between the dimension and similar subscales on previous measures (See Brennan, et al., 1998 for more information). Internal consistency of the ECR for this study was α =.89 for the full scale, α =.94 for the attachment avoidance scale, and α =.91 for the attachment anxiety scale. Posttraumatic Cognitions Inventory (PTCI). The PTCI (Foa et al., 1999) is a 36- item measure of trauma related thoughts and beliefs. Specifically, the PTCI examines negative cognitions about the self, negative cognitions about the world, and self blame. Respondents rate their level of agreement on a 7-point Likert scale (totally disagree to 10

11 PTSD Cognitions & Attachment in Victims 11 totally agree) to a series of statements (e.g., The world is a dangerous place ). Foa and colleagues provide evidence that PTCI scores correlate with PTSD severity (r =.79), depression (r =.75), and anxiety (r =.75). In this study, internal consistency for the full scale was excellent (α =.97), and each of the subscales were acceptable (self, α =.96; world, α =.92; self-blame, α =.78). Purdue PTSD Scale-Revised (PPTS-R). The PPTS-R (Lauterbach & Vrana, 1996) is a 17-item questionnaire that assesses symptoms found within PTSD criteria B, C, and D. Respondents rate the frequency of occurrence within the previous month of each item on a five-point Likert scale ranging from not at all to often. Answers can be scored using a dichotomous index for the presence or absence of PTSD or a continuous measure of severity. The full measure showed excellent internal consistency (α =.95) and the subscales demonstrated good internal consistency (re-experiencing, α =.88; avoidance, α =.88; and arousal, α =.89) in this study. The PPTS-R correlates with other measures of PTSD symptomatology (Lauterbach & Vrana, 1996). The present study uses the PPTS-R as a measure of PTSD symptomatology, but it should be noted that the measure does not assess emotional reactions to the event and therefore is not a complete assessment of the diagnostic criteria. Revised Conflict Tactics Scales (CTS2). The CTS2 (Straus, Hamby, Boneymccoy, & Sugarman, 1996) is a revised version of the CTS1 (Straus, 1979), which has been widely used in both research studies and the assessment and evaluation of trauma victims. The CTS2 is a 78-item measure of violence in romantic relationships. In this study, 39- items assessing physical and sexual victimization were administered. The present study used a restrictive definition of physical abuse, including only the major assault and major 11

12 PTSD Cognitions & Attachment in Victims 12 injury items in the physical trauma category. Similarly, the major sexual coercion items were included as items assessing sexual assault. Respondents report the frequency of events in the previous year on a Likert scale ranging from 1 (once in the past year) to 6 (more than twenty times in the past year), with options of not in the past year but happened before and never happened. Validity evidence on the CTS2 is provided by Strauss and colleagues (Strauss et al., 1996). Internal consistency for the 39 items used in this study was excellent (α =.92). Sexual Experiences Survey-Modified (MSES; Koss & Oros, 1982). The modified SES is a 16-item questionnaire that assesses levels of sexual victimization and sexual harassment. Respondents read descriptions of a variety of situations and indicate if they experienced the event since the age of 14. The survey has shown acceptable reliability (α =.74) and validity when compared to interview responses (r =.73; Koss & Gidycz, 1985). Internal consistency for this study was excellent (α =.90). While the MSES assesses a range of sexual abuse items, only the 8 attempted rape and completed rape items were considered endorsement of sexual assault. Results Although endorsement of the experience of interpersonal violence, as assessed by the MSES and the CTS2, does not guarantee that the event was perceived by the individual as being traumatic and does not sufficiently assess criterion A for PTSD (i.e., personal response to the event), research indicates that the experience of interpersonal trauma is strongly linked with the development of the disorder (Kessler, Sonnega, Bromet, & Hughes, 1995). For the purposes of the study, a dichotomous trauma status variable (trauma or no trauma) was created, based on answers to the interpersonal 12

13 PTSD Cognitions & Attachment in Victims 13 violence questionnaires. Thirty-four percent of the participants (38% of females and 28% of males) endorsed one or more items of interpersonal trauma; fifty-nine percent of participants did not endorse interpersonal trauma; and seven percent failed to complete the questionnaires. Victims of interpersonal trauma endorsed higher levels of anxious (M = 17.83, SD = vs. M = 12.37, SD = 9.03; F Δ = 16.5, p <.01), depressive (M = 12.37, SD = vs. M = 7.99, SD = 7.86; F Δ = 12, p <.01), and PTSD symptoms (M = 18.07, SD = vs. M = 7.37, SD = 13.03; F Δ = 10.24, p <.01) than non-victims in the current sample. Pearson correlations between endorsed trauma status, PTSD related cognitions, attachment cognitions, and psychological symptoms are presented in Table 1. Past research has suggested that sex and minority status may serve as risk factors for some psychological disorders, including PTSD and other anxiety disorders as well as depression (Breslau, Schultz, & Peterson, 1995, Kessler et al., 1995). Although not a primary hypothesis, the two demographic variables assessed in the present study (i.e., race and sex) were controlled for in all analyses to prevent illusory relations between the variables of interest and symptomatology. Multivariate Analyses of Variance (MANOVAs) were conducted to examine the effects of sex, and race on total levels of endorsed PTSD, anxiety, and depression symptoms. The sex of the participant was not a significant predictor of overall levels of psychological symptoms endorsed, F(3,211) =.25, p >.05. However, the race of the participant yielded a significant result, F(12, 639) = 2.00, p <.05. Univariate analyses revealed that race served as a significant predictor of endorsed PTSD symptoms, F(4,213) = 4.63, p <.01, but was not a significant predictor of anxious or depressive symptoms. Participants that identified themselves as Asian reported the highest levels of PTSD symptoms (Asian, M = 25.57, SD = 19.64; 13

14 PTSD Cognitions & Attachment in Victims 14 Caucasian, M = 10.24, SD = 14.96; African American, M = 10.81, SD = 12.15). However, the small number of participants that comprised this group (n = 7) make it difficult to draw any conclusions about the strength of the relationship between Asian American status and the development of PTSD symptoms. In accordance with Baron and Kenny (1986), the proposed moderational models were tested using multiple regression techniques. For each examination of moderation, the independent variable and the moderator were entered in the first step, followed by the interaction of the two variables. Moderator effects are indicated by a significant effect of the interaction term while controlling for the effects of the independent and moderator variables (Baron & Kenny, 1986). Hypothesis 1: Attachment, Trauma, and Psychological Symptoms Insecure attachment cognitions were proposed to moderate the relation between the experience of an interpersonal trauma and the development of psychological symptoms. Prior to the examination of the interaction between trauma and the attachment dimensions, the main effects of trauma on the attachment dimensions were tested. A significant relation was found between trauma status and attachment anxiety with victims endorsing higher levels of attachment anxiety than non-victims (M = 4.05, SD = 1.14 vs. M = 3.76, SD = 1.11; F Δ = 5.01, R 2 Δ =.02, p <.05). A similar regression analysis was performed assessing the influence of trauma status on attachment avoidance, but the relation was non-significant (M = 3.05, SD = 1.20, vs. M = 2.92, SD = 1.21; F Δ =.87, R 2 Δ =.00, p >.05). Given this pattern of relations, attachment anxiety may function as either a vulnerability to or consequence of interpersonal trauma, while attachment avoidance appears to be unrelated to trauma status. 14

15 PTSD Cognitions & Attachment in Victims 15 Multiple regression analyses were used to determine whether symptom development after the experience of interpersonal trauma differed as a function of adult romantic attachment dimensions. For each analysis, demographic information was entered in block one. Trauma status and an attachment dimension were then added in block two and the trauma status X attachment dimension interaction was added in block three. As shown in Tables 2 and 3, significant main effects were found for attachment anxiety, attachment avoidance, and trauma status. Participants who endorsed higher levels of attachment anxiety reported higher levels of each of the psychological symptoms than participants who endorsed low levels of attachment anxiety. Similarly, participants who endorsed higher levels of attachment avoidance reported higher levels of psychological symptoms than participants who endorsed low levels of attachment avoidance. No significant interaction between the experience of trauma and either attachment dimension on psychological symptoms was found. Therefore, that hypothesis that the attachment dimensions would function as a moderator between victimization and the development of psychological symptoms was not supported. Hypothesis 2: PTSD-related cognitions, trauma, and symptom development The second hypothesis predicted that PTSD-related cognitions would moderate the relation between the experience of interpersonal trauma and the development of psychological symptoms. A main effect of trauma on PTSD-related cognitions emerged, indicating that victims endorsed higher levels of PTSD-related cognitions than nonvictims (F Δ = 22.32, R 2 Δ =.08, p <.01). Multiple regression analyses were performed to examine the relations between PTSD-related cognitions, victimization, and psychological symptoms. As shown in Table 4, level of PTSD-related cognitions was significantly 15

16 PTSD Cognitions & Attachment in Victims 16 related to each of the psychological symptoms, with higher levels of PTSD-related cognitions associated with higher levels of each of the psychological symptoms. When the influence of the PTSD-related cognitions was accounted for, trauma status was only related to the development of PTSD symptoms. Significant interactions were found for PTSD-related cognitions and victimization on levels of depressive and PTSD symptoms, but not on level of anxious symptoms. Victims that reported high levels of PTSD-related cognitions reported higher levels of PTSD or depressive symptoms than individuals who endorsed only the presence of trauma, only high PTSD-related cognitions, or neither the experience of trauma nor a high level of PTSD-related cognitions. Moreover, the PTSDrelated cognitions by trauma status interaction was significant on all PTSD symptom clusters. Hypothesis 3: PTSD related cognitions and attachment Hypothesis three predicted that PTSD-related cognitions would moderate the relation between the attachment dimensions and the development of psychological symptoms. As shown in tables 5 and 6, main effects were found for PTSD-related cognitions in all analyses. Main effects remained for attachment anxiety on anxious and depressive symptoms, but not on PTSD symptoms, when PTSD-related cognitions were entered simultaneously. However, attachment avoidance was not significantly associated with any of the psychological symptoms when entered simultaneously with PTSD-related cognitions. The attachment anxiety by PTSD-related cognition interaction was only significant in predicting depressive symptoms. Participants that endorsed high levels of both attachment anxiety and PTSD-related cognitions endorsed higher levels of 16

17 PTSD Cognitions & Attachment in Victims 17 depressive symptoms than those high in either attachment anxiety or PTSD-related cognitions or low in both. See Figure 1 for a graphic depiction of the PTSD-related cognition and attachment anxiety interaction. The three way interactions between PTSDrelated cognitions, trauma status, and each of the adult romantic attachment dimensions were non-significant on all psychological symptoms. Discussion Despite the fact that the college student population is generally considered high functioning and advantaged compared to peers, high percentages of interpersonal trauma are reported in student populations (Green et al., 2000). The present study found that 38% of the female and 28% of the male college student sample endorsed at least one instance of adult interpersonal trauma. Consistent with previous findings (Kozaric-Kovacic, 2001; Orsillo et al., 1996; Zayfert et al., 2002), victims of interpersonal trauma demonstrated higher levels of anxious, depressive, and PTSD symptoms than non-victims. These findings reinforce the importance of studying interpersonal trauma and its psychological sequelae in college students. Adult Romantic Attachment, Trauma, and Psychological Symptoms Insecure attachments have been suggested as one of the consequences of abuse in childhood (Alexander, 1992; Muller et al., 2000). This study extends previous literature on attachment and interpersonal trauma by examining victims of adult interpersonal trauma and employing a dimensional approach to assessing romantic attachment. Victims in this study demonstrated higher levels of attachment anxiety than non-victims but did not display higher levels of attachment avoidance. These findings suggest that victims are not more likely to avoid intimacy than their non-victim peers. Rather, the 17

18 PTSD Cognitions & Attachment in Victims 18 higher levels of attachment anxiety suggest that victims may find it more difficult to feel secure and to trust their significant others once they are in relationships. Replication of this pattern would indicate that victims may benefit from therapy and/or skills training addressing the maintenance of relationships rather than engagement in relationships. In the present study, both attachment anxiety and attachment avoidance had significant main effects on levels of anxious, depressive, and PTSD symptoms, in which participants that endorsed higher levels on the attachment dimensions displayed higher levels of symptoms. None of the interactions between trauma status and either attachment dimension were significant for anxiety, depression, or PTSD. Results were consistent with previous studies which indicate that insecure attachment is related to the development of psychological symptoms (Muller et al., 2000), but failed to demonstrate a unique relationship between adult romantic attachment and interpersonal trauma. PTSD-related Cognitions The Posttraumatic Cognitions Inventory (PTCI; Foa et al., 1999) was created to assess cognitions believed to be involved in the development and maintenance of PTSD. The cognitions were selected based on theory of the development of PTSD and clinical experience with trauma survivors. The cognitive beliefs were designed to assess cognitions specifically associated with the traumatic event and changes in the individual as a result of the experience. If this is the case, PTSD related cognitions should be associated with psychological symptoms in general, should be strongly associated with PTSD, and should presumably be a better predictor of PTSD than other more general vulnerabilities, such as attachment cognitions. Consistent with Foa and colleagues (1999) findings, higher levels of PTSD-related cognitions were significantly related to 18

19 PTSD Cognitions & Attachment in Victims 19 higher levels of anxious, depressive, and PTSD symptoms in the current study. In addition, an interaction between trauma status and PTSD-related cognitions was found predicting PTSD symptoms and depressive symptoms. Specifically, participants in the trauma group with low levels of PTSD related cognitions and participants in the nontrauma group with high levels of PTSD-related symptoms demonstrated similar levels of symptoms, while participants who previously experienced trauma and reported high levels of PTSD-related cognitions reported the highest levels of symptoms. The findings that PTSD related cognitions significantly interacted with trauma status to predict endorsed level of PTSD symptoms while neither of the attachment dimensions yielded a significant interaction with trauma status supports PTSD related cognitions as a unique predictor of PTSD. This set of PTSD-related cognitions may function as a useful tool for assessing both risk for PTSD and progress in treatment. Attachment and PTSD-related Cognitions Adult romantic attachment dimensions were predicted to moderate between the levels of PTSD-related cognitions and overall level of symptoms. This moderational model was supported for the development of depressive symptoms, but not for anxious or PTSD symptoms. In addition to the vulnerability to the development of symptoms one experiences by possessing high attachment cognitions or high PTSD-related cognitions, individuals with high levels of both cognitions may experience an increased vulnerability to the development of depressive symptoms. It is proposed that individuals with high levels of both PTSD related cognitions and attachment cognitions are at an increased risk for the development of depressive symptoms due to the negative beliefs possessed by the individuals regarding the self, the world, and the dependability of others. It seems likely 19

20 PTSD Cognitions & Attachment in Victims 20 that globally held negative beliefs would be linked with feelings of hopelessness and therefore depressive symptoms. Therefore individuals with high levels of negative beliefs in multiple domains would be at increased risk for the development of depressive symptoms. The lack of a significant interaction between attachment and PTSD related cognitions predicting PTSD could be due to strength of the association between the PTSD related cognitions and the development of PTSD symptoms. It is possible that the attachment cognitions are not associated strongly enough with the development of PTSD symptoms to contribute variance accounted for above and beyond the PTSD related cognitions. The simultaneous addition of the adult romantic attachment dimensions and the PTSD-related cognitions in the regression equations allowed the independent influences to be examined. When the effect of attachment avoidance cognitions were examined in conjunction with the effect of the PTSD-related cognitions, only the level of PTSDrelated cognitions was a significant predictor of the psychological symptoms. When attachment anxiety and PTSD-related cognitions were examined simultaneously both remained predictors of anxious and depressive symptoms, but attachment anxiety did not predict level of PTSD symptoms (PTSD-related cognitions remained a significant predictor). While previous studies have established a link between insecure attachment and PTSD (Muller et al., 2000); this study suggests that PTSD-related cognitions might be a better predictor of PTSD symptoms than attachment avoidance or attachment anxiety. The current study has several limitations. While interpersonal trauma is one of the largest precursors to the development of PTSD, many other traumatic events also result in 20

21 PTSD Cognitions & Attachment in Victims 21 the development of PTSD and other psychological disorders. It is likely that some of the individuals in the non-victim group experienced a traumatic event that was not assessed or not endorsed and that the experience influenced their responding. In addition, the present study did not assess the participants reaction to the interpersonal violence or the event that led to the endorsed post-trauma reactions. The exclusive use of college students and the reliance on self-report measures in the present study are additional limitations. Confidence in the results of the analyses could be increased if the sample size were larger. Follow-up studies larger sample sizes are needed. Finally, the crosssectional design of this study is a second limitation, as it prevents examinations of the direction of causality between the psychological variables in the present study. The present study examined several relations between proposed vulnerabilities and psychological symptoms developed subsequent to adult interpersonal violence using a moderational approach. Past research has found support for both mediational (Roche et al., 1999; Shapiro & Levendosky, 1999; Twaite & Rodriguez-Srednicki, 2004) and moderational (Whiffen et al., 1999) models. Moderational models were chosen for the present study due to the lack of knowledge about the temporal relations of the variables. The cognitive styles of the victims in this study were not assessed before the event, and thus it is unknown whether they function as risk factors for experiencing interpersonal trauma, previously developed cognitive vulnerabilities, or reflect a change in beliefs after the experience of the traumatic event. A longitudinal design would be necessary to examine the direction of causality that exists between victim experience and insecure attachment or PTSD-related cognitions. It may be beneficial for future studies that are 21

22 PTSD Cognitions & Attachment in Victims 22 able to assess these vulnerabilities longitudinally to test the relations using a mediational approach. Conclusions This study extended extant research by examining the contribution of one risk factor (i.e., trauma status) and two vulnerability factors (i.e., adult romantic attachment, and PTSD-related cognitions) for anxiety, depression, and PTSD. Trauma status served as an apparent risk factor for the development of anxiety, depression, and PTSD. Both attachment anxiety and PTSD-related cognitions are suggested as vulnerability factors for the development of anxious and depressive symptoms. In addition, PTSD-related cognitions are suggested as a unique vulnerability factor for the development of PTSD in victims of interpersonal trauma. While the interpersonal aspects of both attachment and PTSD-related cognitions suggest that the two may be linked, the moderational model was only supported for the development of depressive symptoms. Future research examining more specific trauma groups and using longitudinal designs is needed. 22

23 PTSD Cognitions & Attachment in Victims 23 References Alexander, P. C. (1992). Application of attachment theory to the study of sexual abuse. Journal of Consulting and Clinical Psychology, 60, Baron, R. M., & Kenny, D. A. (1986). The moderator-mediator variable distinction in social psychological research: Conceptual, strategic, and statistical considerations. Journal of Personality and Social Psychology, 51, Beck, A.T., Epstein, N., Brown, G., & Steer, R. A. (1988). An inventory for measuring clinical anxiety: Psychometric properties. Journal of Consulting and Clinical Psychology, 56, Beck, A.T., Steer, R. A., & Brown, G. K. (1996). Manual for the Beck Depression Inventory-II. San Antonio, TX: Psychological Corp. Beck, A.T., Steer, G., Garbin, M. (1988). Psychometric properties of the Beck Depression Inventory: 25 years of evaluation. Clinical Psychology Review, 8, Bowlby, J. (1973). Attachment and loss: Vol. 2 Separation: Anxiety and Anger. New York: Basic Books. Brennan, K. A., Clark, C. L., & Shaver, P. R. (1998). Self-report measurement of adult attachment: An integrative overview. In Simpson & Roles, Attachment theory and close relationships (pp ). New York: Guilford Press. Breslau, N., Schultz, L., Peterson, E. (1995). Sex difference in depression: A role for preexisting anxiety. Psychiatry Research, 58,

24 PTSD Cognitions & Attachment in Victims 24 Carlier, I. V., Gersons, B. P. R. (1995). Partial Posttraumatic stress disorder (PTSD): The issue of psychological scars and the occurrence of PTSD symptoms. Journal of Nervous & Mental Disease, 183(2), Cassidy, J. (1995). Attachment and Generalized Anxiety Disorder. In: D. Cicchetti and S. L. Toth (Eds.), Emotion, cognition, and representation (pp ). Rochester, NY, US: University of Rochester Press. Collins, N. L., & Read, S. J. (1990). Adult attachment, working models, and relationship quality in dating couples. Journal of Personality and Social Psychology, 67, Ehlers, A., & Clark, D. M. (2000). A cognitive model of posttraumatic stress disorder. Behavior Research and Therapy, 38, Foa, E. B., Ehlers, A., Clark, D. M., Tolin, D. F., & Orsillo, S. M. (1999). The posttraumatic cognitions inventory (PTCI): development and validation. Psychological Assessment, 11, Foa, E. B., Stekette, G., & Rothbaum, B. O. (1989). Behavioral/cognitive conceptualizations of posttraumatic stress disorder. Behavior Therapy, 20, Foa, E. B., & Riggs, D. S. (1993). Posttraumatic stress disorder in rape victims. In Oldham, J., Riba, M.B., & Tasman, A. (Eds.). American psychiatric press review of psychiatry, Vol. 12. (pp ). Washington, D.C.: American Psychiatric Press. Foa, E. B., & Rothbaum, B. O. (1998). Treating the trauma of rape: Cognitive behavioral therapy for PTSD. New York: Guilford Press. 24

25 PTSD Cognitions & Attachment in Victims 25 Green, B. L., Goodman, L. A., Krupnick, J. L., Corcoran, C. B., Petty, R. M., Stockton, P., & Stern, N. M. (2000). Outcomes of single versus multiple trauma exposure in a screening sample. Journal of Traumatic Stress, 13(2), Griffin, D., & Bartholomew, K. (1994). The metaphysics of measurement: The case of adult attachment. In K. Bartholomew & D. Perlman (Eds.), Advances in personal relationships: Vol. 5. Attachment processes in adulthood (pp ). London: Jessica Kingsley. Haaga, D. F., Yarmus, M., Hubbard, S., Brody, C., Solomon, A., Kirk, L., Chamberlain, J. (2002). Mood dependency of self-rated attachment style. Cognitive Therapy and Research, 26, Hammen, C. L., Burge, D., Daley, S. E., Davilia, J., Paley, B., & Rudolph, K. D. (1995). Interpersonal attachment cognitions and predictions of symptomatic responses to interpersonal stress. Journal of Abnormal Psychology, 104(3), Kessler, R. C., Sonnega, A., Bromet, E., Hughes, M., & Nelson, C. (1995). Posttraumatic stress disorder in the National Comorbidity Survey. Archives of General Psychiatry, 52, Kirk, A., & Dollar, S. A. (2002). Prevalence of traumatic event sand PTSD symptomatology among a selected sample of undergraduate students. Journal of Social Work in Disability and Rehabilitation, 1, Koss, M. P., & Gidycz, C. A. (1985). Sexual experiences survey: Reliability and validity. Journal of Consulting and Clinical Psychology, 53(3),

26 PTSD Cognitions & Attachment in Victims 26 Koss, M. P., & Oros, C. J. (1982). Sexual experiences survey: A research instrument investigating sexual aggression and victimization. Journal of Consulting and Clinical Psychology, 50(3), Kozaric-Kovacic, D. (2001). Assessment of posttraumatic stress disorder and comorbidity. Military Medicine, 166(8), Lauterbach, D., & Vrana, S. (1996). Three studies on the reliability and validity of a selfreport measure of posttraumatic stress disorder. Assessment, 3, Main, M., Kaplan, N., & Cassidy, J. (1985). Security in infancy, childhood, and adulthood: A move to the level of representation. Monographs of the Society for the Research in Child Development, 50, McCarthy, G., & Taylor, A. (1999). Avoidant/ambivalent attachment style as a mediator between abusive childhood experiences and adult relationship difficulties. Journal of Child Psychology and Psychiatry, 40(3), Muller, R. T., & Lemieux, K. E. (2000). Social support, attachment, and psychopathology in high risk formerly maltreated adults. Child Abuse & Neglect, 24, Muller, R. T., Sicoli, L. A., & Lemieux, K. E. (2000). Relationship between attachment style and posttraumatic stress symptomatology among adults who report the experience of childhood abuse. Journal of Traumatic Stress, 13(2), Orsillo, S. M., Weathers, F. W., Litz, B. T., Steinberg, H. R., Huska, J. A., & Keane, T. M. (1996). Current and lifetime psychiatric disorders among veterans with war zone-relate posttraumatic stress disorder. Journal of Nervous and Mental Disease, 184,

27 PTSD Cognitions & Attachment in Victims 27 Resick, P.A. (1993). The psychological impact of rape. Journal of Interpersonal Violence, 8(2), Roche, D. N., Runtz, M. G., Hunter, M. A. (1999). Adult attachment: A mediator between child sexual abuse and later psychological adjustment. Journal of Interpersonal Violence, 14, Scarpa, A., Fitkretoglu, D., Bowser, F., Hurley, J. D., Pappert, C. A., Romero, N., Van Voorhees, E. (2002). Community violence exposure in university students. Journal of Interpersonal Violence, 17, Shapiro, D. L., & Levendosky, A. A. (1999). Adolescent survivors of childhood sexual abuse: The mediating role of attachment style and coping in psychological and interpersonal functioning. Child Abuse & Neglect, 23, Simpson, J. A. (1990). Influence of attachment styles on romantic relationships. Journal of Personality and Social Psychology, 59, Sternberg, K. J., Lamb, M. E., Guterman, E., Abbott, C. B., Dawud-Noursi, S. (2005). Adolescents perceptions of attachments to their mothers and fathers in families with histories of domestic violence: A longitudinal perspective. Child Abuse & Neglect, 29, Strauss, M. A. (1979). Measuring intrafamily conflict and violence: The conflict tactics scale. Journal of Marriage and Family, 41, Strauss, M. A., Hamby, S. L., Boneymccoy, S., & Sugarman, D. B. (1996). The revised conflict tactics scales (Cts2) development and preliminary psychometric data. Journal of Family Issues, 17(3),

28 PTSD Cognitions & Attachment in Victims 28 Twaite, J. A., & Rodriguez-Srednicki, O. (2004). Childhood sexual and physical abuse and adult vulnerability to PTSD: The mediating effects of attachment and dissociation. Journal of Child Sexual Abuse, 13, Vrana, S., & Lauterbach, D. (1994). Prevalence of traumatic events and post-traumatic psychological symptoms in a nonclinical sample of college students. Journal of Traumatic Stress, 7, Whiffen, V.E., Judd, M.E., & Aube, J.A. (1999). Intimate relationships moderate the association between childhood sexual abuse and depression. Journal of Interpersonal Violence, 14, Williams, N.L., & Riskind, J. (2004). Adult romantic attachment and cognitive vulnerabilities to anxiety and depression: Examining the interpersonal basis of vulnerability models. Journal of Cognitive Psychotherapy, 18, Zayfert, C., Dums, A. R., Ferguson, R. J., & Hegel, M. T. (2002). Health functioning impairments associated with posttraumatic stress disorder, anxiety disorders, and depression. Journal of Nervous and Mental Disease, 190(4),

29 PTSD Cognitions & Attachment in Victims 29 Table 1. Pearson Correlations between Trauma Status, PTSD Related Cognitions, Attachment Cognitions, and Psychological Symptoms BAI total.65**.48**.18**.33**.51**.25** 2. BDI total.50**.26**.38**.65**.23** 3. PTSD total.18**.19**.46**.33** 4. Attachment Avoidance.03.37** Attachment Anxiety.33**.12* 6. PTCI Total.28** 7. Trauma Status Note. BAI = Beck Anxiety Inventory, BDI = Beck Depression Inventory, PTCI = Posttraumatic Cognitions Inventory. * = p <.05, ** = p <

30 PTSD Cognitions & Attachment in Victims 30 Table 2. Summary of the Regression Analyses Examining the Relationship between Attachment Anxiety, Trauma, and the Psychological Symptoms Symptom Block F R 2 Sig. Variable in Beta Measure FΔ R 2 Δ Block PPTSD 1. Demographics 7.78** 0.06 Race.24** 2. Attach Anx 17.91** 0.13 Attach Anx 0.18** Trauma Trauma 0.29** 3. AnxTrauma BAI 1. Demographics 3.29* 0.03 Sex 0.05* 2. Attach Anx 24.05** 0.16 Attach Anx 0.34** Trauma Trauma 0.17** 3. AnxTrauma BDI 1. Demographics Attach Anx 27.70** 0.18 Attach Anx 0.38** Trauma Trauma 0.13* 3. AnxTrauma Note. F and R 2 values are reported for the first block in each regression equation. FΔ and R 2 Δ values are reported for the second and third blocks of each regression equation. * = p <.05, ** = p <.01. PPTSD = Purdue PTSD Scale- Revised, BAI = Beck Anxiety Inventory, and BDI = Beck Depression Inventory, Trauma = assessment of interpersonal trauma, Attach Anx = Attachment Anxiety, AnxTrauma = the interaction term for the interaction between Attachment Anxiety and Trauma 30

31 PTSD Cognitions & Attachment in Victims 31 Table 3. Summary of the Regression Analyses Examining the Relationship between Attachment Avoidance, Trauma, and the Psychological Symptoms Symptom Block F/ R 2 / Significant Beta Measure FΔ R 2 Δ Variable in Block PPTSD 1. Demographics 7.78** 0.06 Race.24** 2. Attach Avoid 15.69** 0.11 Attach Avoid 0.13* Trauma Trauma 0.30** 3. AvoidTrauma BAI 1. Demographics 3.29* 0.03 Sex 0.16* 2. Attach Avoid 10.09** 0.07 Attach Avoid 0.17** Trauma Trauma 0.21** 3. AvoidTrauma BDI 1. Demographics Attach Avoid 13.86** 0.10 Attach Avoid 0.26** Trauma Trauma 0.17** 3. AvoidTrauma Note. F and R 2 values are reported for the first block in each regression equation. FΔ and R 2 Δ values are reported for the second and third blocks of each regression equation. * = p <.05, ** = p <.01. PPTSD = Purdue PTSD Scale- Revised, BAI = Beck Anxiety Inventory, and BDI = Beck Depression Inventory, Trauma = assessment of interpersonal trauma, Attach Avoid = Attachment Avoidance, AvoidTrauma = the interaction term for the interaction between Attachment Avoid and Trauma 31

32 PTSD Cognitions & Attachment in Victims 32 Table 4. A Summary of the Regression Analyses Examining the Relations between PTSD-related Cognitions, Victimization, and the Psychological Symptoms Symptom Block F/ R 2 / Significant Beta Measure FΔ R 2 Δ Variable in Block PPTSD 1. Demographics 7.19** 0.06 Race 0.23** 2. PTSD Cog 41.90** 0.25 PTSD Cog 0.40** Trauma Trauma 0.21** 3. PTCTrauma 17.19** 0.05 PTCTrauma 0.64** BAI 1. Demographics 3.42* 0.03 Sex 0.16** 2. PTSD Cog 47.31** 0.27 PTSD Cog 0.48** Trauma 3. PTCTrauma BDI 1. Demographics 3.31* 0.03 Sex 0.16** 2. PTSD Cog 99.90** 0.43 PTSD Cog 0.65** Trauma 3. PTCTrauma 4.21* 0.01 PTCTrauma 0.28* Note. F and R 2 values are reported for the first block in each regression equation. FΔ and R 2 Δ values are reported for the second and third blocks of each regression equation. * = p <.05, ** = p <.01. PPTSD = Purdue PTSD Scale- Revised, BAI = Beck Anxiety Inventory, and BDI = Beck Depression Inventory, Trauma = assessment of interpersonal trauma, PTSD Cog = PTSD related cognitions, PTCTrauma = the interaction term for the interaction between trauma and PTSD related cognitions. 32